Last week I spent a few days at Harvard Kennedy School for the Public Interest Tech Summit, where I joined dozens of tech fellows and scholars from various backgrounds to explore and discuss around the concept of “public interest technology.” If you are interested, you can click here to watch the video recording of the opening session. A couple of ideas and inspirations I would like to share in the following.
“Public interest technology” becomes increasingly important and imperative today. The world is technological – technology is everywhere, revolutionizing every industry, transforming how the society operates, and reshaping how we work, live and relate. Along with the changes emerge many new issues like systematic biases by algorithm, data security threats, and outdated laws. While people can vote for government leaders, they have no influence on the private tech giants who also exert great influences on their lives through the technology products and services. A more transparent and responsible mechanism is needed to ensure that technology will make our life better not the opposite.
Data policy plays an important role in this mechanism. I noticed a strong interest among the people I met at HKS attempting to figure out the role of data in today’s society. Some said data is the new oil of the digital era, because data fuels a wide range of industries and can provide additional information by “data mining.” Some others thought data should be deemed as infrastructure, essential to operation of today’s society and thus should be used for public good.
There are criticisms against both analogies, as a quick google search will tell. But I would like to focus on their merits.
Firstly, data is a valuable and crucial resource, but neither only extractive technology can make data useful, nor the value of data should only serve the interests of those who control the data. Data provides constantly evolving knowledge necessary to inform better practices. Within the framework of respecting privacy rights, data should be wildly available and used for public good. Particular alert on data colonialism. Meaningful access and participation regarding the related data should be given to people of the data origin. After all, it is their data, and they should have the right to benefit from their data.
Secondly, considering the fundamental significance of data for people’s lives and the normal function of a society, a fiduciary duty should impose on data holders and follow the full life cycle of data. As we have already seen, asymmetrical relationships largely exist between individuals and data holders. Most algorithms are still operated in a “black box,” as the book The Black Box Society shows. And many algorithms, even if being transparent and explainable, are too complicated for many people to understand. Protecting individuals from being harmed by their own data requires a responsible manner in data processing. Data should be collected responsibly, bias in the dataset should be eliminated, and abusive usage of data should not be allowed. In addition, governments should design the regulation of data to ensure the marginalized communities would not be adversely impacted, and data cannot be used in a way to further widen the gap separating the society.
